On Wed, Jun 09, 1999 at 02:46:29AM +0100, Andrew Collier wrote: > >If identifiable sections of that work are not derived from the Program, > >and can be reasonably considered independent and separate works in > >themselves, then this License, and its terms, do not apply to those > >sections when you distribute them as separate works.
> Okay, I've just noticed that they indeed do have to be distributed > seperately, but that should then fit the terms. Nope. Not only do they need to be distributed separately, but they also need to be considered independent works in themselves. A package calling itself "The Win32 version of Simcoupe - by the way you need SimCoupe to use this" hardly fits that description. > >Whatever it is I don't think "loosely defined" could ever be applied to it. > In that case, where does it define "distribution"? Does it count, for > example, if a few copies of the binary have been given to beta testers but > that the source has not? OK, this may be a grey area (but one grey area does not make the whole thing "loosely defined"). But giving it to a few selected beta testers counts as part of the development rather than distribution, I would say. > Better licenses give the original programmer magic extra rights which are > not passed on to subsequent developers, ie that only the original > programmer can produce an official distribution. > But consider A, who writes a program called Foo v1.0, released GPL. > A continues development, and releases Foo v1.1, released GPL. > B develops from Foo v1.0 source, and releases Foo v1.1 - which is now > totally incompatible with A's program. But note that B's version contains, in accordance with paragraph 2(a), prominent notices that B has carried out the modifications. If A thinks this is likely to be a problem he is perfectly free to amend the licence under which he distributes the program. > PS. This is getting *way* off topic. Can we do something more appropriate > with it? It's been a while since anyone posted to oxbridge.tat; perhaps... You first. imc
